Vehicle Travel Control Using Curvature-Based Line Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing travel control devices for autonomous vehicles face challenges in maintaining accurate line recognition, particularly due to peak phenomena and non-uniform travel environments, which can lead to decreased performance and inaccuracy in vehicle control.
Innovation Solution
The proposed solution involves a compensation algorithm that adjusts line recognition results based on the curvature and curvature change rate of the road, identifying a peak phenomenon and generating calibrated line information using an expected curvature change rate. This algorithm selectively compensates for line recognition inaccuracies and ensures continuous accurate vehicle control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a polynomial-based line recognition algorithm is used to identify road lines, then the travel control device can derive line information, but accuracy momentarily decreases due to peak phenomena in polynomial coefficients
Solution Approach 1:
The system performs preliminary detection of peak phenomena in polynomial coefficients before they significantly degrade line recognition accuracy. By detecting curvature change rates and identifying peak conditions in advance, the system can prepare compensatory measures to maintain continuous accurate control
Solution Approach 2:
The system continuously monitors polynomial coefficients and curvature change rates, providing feedback to detect peak phenomena. When peaks are detected, the system adjusts line recognition results based on the magnitude and direction of curvature changes, creating a closed-loop control system that maintains accuracy despite coefficient fluctuations
2Adaptability or versatility
If the travel control device relies on sensor data for line recognition, then it can adapt to road conditions, but accuracy decreases when the travel environment changes or when outside the sensor's effective measuring range
Solution Approach 1:
The system prepares for potential accuracy degradation by continuously monitoring curvature change rates and detecting peak phenomena before they cause significant errors. This proactive approach cushions against accuracy loss by identifying problematic conditions early, allowing the system to switch to alternative recognition methods or adjust parameters before sensor limitations significantly impact performance
Solution Approach 2:
The system uses curvature change rate analysis as an intermediary mechanism to bridge sensor data and line recognition results. By analyzing the rate of change in curvature, the system can detect anomalies and compensate for sensor limitations, acting as a mediator that maintains accuracy even when direct sensor measurements become unreliable
3Measurement precision
If the system continuously monitors and compensates for line recognition results, then accuracy is maintained, but computational complexity and processing time increase
Solution Approach 1:
The system applies compensation algorithms selectively rather than uniformly. By detecting peak phenomena in specific polynomial coefficients and applying compensation only when and where needed, the system maintains high accuracy while minimizing unnecessary computational overhead. The compensation is localized to specific segments of the line recognition process where peak phenomena occur
Data Source
AI summary
Disclosed is a device for controlling travel. The device includes a sensor, a memory, and a controller. For example, the device may obtain, via a sensor (e.g., a LIDAR), a line recognition result associated with a road on which a vehicle is traveling, determine, based on the line recognition result, whether information, which comprises at least one of a curvature of the road or a curvature change rate of the road, satisfies a specified condition, and, based on the information satisfying the specified condition, generate calibrated line information using an expected curvature change rate, wherein the expected curvature change rate is determined based on: the curvature change rate, and at least one of an expected heading angle of the vehicle at a target point on the road, an expected curvature at the target point, or an expected lateral error at the target point.